Results of the different analyses performed with the data of the RCT in which corticosteroid injections and physiotherapy were compared with wait and see policy for lateral epicondilytis
| Odds ratio/hazard ratio | 95% CI | p Value | |
|---|---|---|---|
| Naive techniques | |||
| Logistic regression (odds ratio) | |||
| Injection | 0.52 | 0.21 to 1.25 | 0.14 |
| Physiotherapy | 1.97 | 0.67 to 5.82 | 0.22 |
| Traditional Cox regression (hazard ratio) | |||
| Injection | 3.92 | 2.60 to 5.91 | <0.01 |
| Physiotherapy | 1.42 | 0.98 to 2.05 | 0.06 |
| Longitudinal techniques | |||
| Cox regression for recurrent events (hazard ratio) | |||
| Counting approach | |||
| Injection | 1.37 | 1.17 to 1.59 | <0.01 |
| Physiotherapy | 1.21 | 1.03 to 1.43 | 0.02 |
| Total time approach | |||
| Injection | 1.38 | 1.18 to 1.61 | <0.01 |
| Physiotherapy | 1.21 | 1.03 to 1.43 | 0.02 |
| Time to event approach | |||
| Injection | 1.73 | 1.42 to 2.11 | <0.01 |
| Physiotherapy | 1.31 | 1.06 to 1.62 | 0.01 |
| GEE analysis (odds ratio) | |||
| Injection | 2.88 | 1.93 to 4.30 | <0.01 |
| Physiotherapy | 1.63 | 1.10 to 2.43 | <0.01 |
| Random coefficient analysis (odds ratio) | |||
| Injection | 6.01 | 3.73 to 9.70 | <0.01 |
| Physiotherapy | 1.84 | 1.16 to 2.94 | <0.01 |